Chart Visual Question Answering on ChartBench 1.0 (test)
94.33Accuracy (Annotated)GPT 4o
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| GPT 4oPrompting style=Best of Zero-Shot and CoT2025.10 | 94.33 | 36.15 | 52.5 | 91 | 54.53 | |
| ChartAgentPrompting style=Best of Zero-Shot and CoT2025.10 | 94.33 | 60.81 | 70.91 | 91 | 71.39 | |
| DeepSeek-VL2Prompting style=Best of Zero-Shot and CoT2025.10 | 90.75 | 30.31 | 50.28 | 33.5 | 49.39 | |
| Gemini 1.5Prompting style=Best of Zero-Shot and CoT2025.10 | 89.72 | 27.27 | 46.69 | 53.85 | 47.08 | |
| LLama3.2Prompting style=Best of Zero-Shot and CoT2025.10 | 87.58 | 36.38 | 52.22 | 50 | 52.11 | |
| Phi3-visionPrompting style=Best of Zero-Shot and CoT2025.10 | 86.92 | 40.77 | 55.89 | 52 | 55.32 | |
| GPT 4o-miniPrompting style=Best of Zero-Shot and CoT2025.10 | 84.83 | 25.19 | 41.5 | 89.5 | 44.03 | |
| Claude 3 HaikuPrompting style=Best of Zero-Shot and CoT2025.10 | 84.58 | 26.04 | 42.94 | 73 | 44.53 | |
| Qwen2-VLPrompting style=Best of Zero-Shot and CoT2025.10 | 78.42 | 43.5 | 52.94 | 83 | 54.53 | |
| TinyChartPrompting style=Best of Zero-Shot and CoT2025.10 | 77.33 | 32.77 | 47.86 | 28.5 | 46.84 | |
| ChartGemmaPrompting style=Best of Zero-Shot and CoT2025.10 | 75.92 | 22.42 | 39.56 | 35 | 39.32 | |
| InternVL3Prompting style=Best of Zero-Shot and CoT2025.10 | 72.67 | 30.92 | 43.39 | 57 | 44.11 | |
| DePlotPrompting style=Best of Zero-Shot and CoT2025.10 | 70.08 | 28.15 | 39.33 | 78.5 | 41.39 | |
| CogAgentPrompting style=Best of Zero-Shot and CoT2025.10 | 69.92 | 11.62 | 30.28 | 27 | 30.03 | |
| DocOwl1.5Prompting style=Best of Zero-Shot and CoT2025.10 | 67.5 | 23.58 | 37.06 | 44.5 | 37.45 | |
| PixtralPrompting style=Best of Zero-Shot and CoT2025.10 | 66.58 | 28.73 | 39.53 | 63.5 | 40.5 | |
| CogVLMPrompting style=Best of Zero-Shot and CoT2025.10 | 64.83 | 11.62 | 29.03 | 21.5 | 28.42 | |
| ChartVLMPrompting style=Best of Zero-Shot and CoT2025.10 | 61 | 23.92 | 36.97 | 11.5 | 35.63 | |
| MatChaPrompting style=Best of Zero-Shot and CoT2025.10 | 59.5 | 9.69 | 25.86 | 17.5 | 25.42 | |
| OneChartPrompting style=Best of Zero-Shot and CoT2025.10 | 56.78 | 26.81 | 35.22 | 62.76 | 36.81 | |
| ChartInstructPrompting style=Best of Zero-Shot and CoT2025.10 | 55.17 | 20.19 | 31.75 | 22 | 31.24 | |
| UniChartPrompting style=Best of Zero-Shot and CoT2025.10 | 53.5 | 15.96 | 27.44 | 34.5 | 27.82 | |
| SmolVLMPrompting style=Best of Zero-Shot and CoT2025.10 | 47.75 | 14.46 | 23.14 | 58 | 24.97 | |
| ChartLlamaPrompting style=Best of Zero-Shot and CoT2025.10 | 38.25 | 11.42 | 18.81 | 39.5 | 19.89 | |
| SPHINX-VPrompting style=Best of Zero-Shot and CoT2025.10 | 35.91 | 12.3 | 18.08 | 0.5 | 19.76 | |
| Llava1.6Prompting style=Best of Zero-Shot and CoT2025.10 | 35.58 | 9.92 | 16.69 | 42 | 18.03 | |
| mPLUG-Owl3Prompting style=Best of Zero-Shot and CoT2025.10 | 31.08 | 12.65 | 16.92 | 46.5 | 18.47 | |
| Qwen-VL-ChatPrompting style=Best of Zero-Shot and CoT2025.10 | 27.17 | 6.54 | 12.61 | 21 | 13.05 | |
| Llava1.5Prompting style=Best of Zero-Shot and CoT2025.10 | 26.75 | 7 | 13.06 | 16.5 | 13.24 | |
| LlaVA-OneVisionPrompting style=Best of Zero-Shot and CoT2025.10 | 13.25 | 10.5 | 9.94 | 37 | 11.37 | |
| VisualGLMPrompting style=Best of Zero-Shot and CoT2025.10 | 4.83 | 7.65 | 3.92 | 58 | 6.76 | |
| InstructBLIPPrompting style=Best of Zero-Shot and CoT2025.10 | 3.92 | 5.92 | 4.22 | 24.5 | 5.29 | |
| BLIP-2Prompting style=Best of Zero-Shot and CoT2025.10 | 3.67 | 2.92 | 3.11 | 4 | 3.16 |